line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Plucene::SearchEngine::Index; |
2
|
1
|
|
|
1
|
|
848
|
use Carp; |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
83
|
|
3
|
1
|
|
|
1
|
|
5
|
use strict; |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
32
|
|
4
|
1
|
|
|
1
|
|
21
|
use warnings; |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
43
|
|
5
|
1
|
|
|
1
|
|
3264
|
use Module::Pluggable (require => 1, search_path => [qw/Plucene::SearchEngine::Index/]); |
|
1
|
|
|
|
|
13209
|
|
|
1
|
|
|
|
|
9
|
|
6
|
1
|
|
|
1
|
|
97
|
use File::Spec::Functions qw(catfile); |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
62
|
|
7
|
1
|
|
|
1
|
|
986
|
use Plucene::Index::Writer; |
|
1
|
|
|
|
|
161879
|
|
|
1
|
|
|
|
|
36
|
|
8
|
1
|
|
|
1
|
|
1075
|
use UNIVERSAL::require; |
|
1
|
|
|
|
|
1796
|
|
|
1
|
|
|
|
|
12
|
|
9
|
|
|
|
|
|
|
our $VERSION = "1.1"; |
10
|
|
|
|
|
|
|
|
11
|
|
|
|
|
|
|
__PACKAGE__->plugins; |
12
|
|
|
|
|
|
|
|
13
|
|
|
|
|
|
|
=head1 NAME |
14
|
|
|
|
|
|
|
|
15
|
|
|
|
|
|
|
Plucene::SearchEngine::Index - A higher level abstraction for Plucene |
16
|
|
|
|
|
|
|
|
17
|
|
|
|
|
|
|
=head1 SYNOPSIS |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
my $indexer = Plucene::SearchEngine::Index->new( |
20
|
|
|
|
|
|
|
dir => "/var/lib/plucene" |
21
|
|
|
|
|
|
|
); |
22
|
|
|
|
|
|
|
|
23
|
|
|
|
|
|
|
my @documents = map { $_->document } |
24
|
|
|
|
|
|
|
Plucene::SearchEngine::Index::File->examine("foo.html"); |
25
|
|
|
|
|
|
|
|
26
|
|
|
|
|
|
|
$indexer->index($_) for @documents; |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
=head1 DESCRIPTION |
29
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
This module makes it easy to write to Plucene indexes. It does so by |
31
|
|
|
|
|
|
|
providing an interface to the index writer which, in terms of |
32
|
|
|
|
|
|
|
complexity, sits between C and |
33
|
|
|
|
|
|
|
C; it also provides a framework of modules for turning |
34
|
|
|
|
|
|
|
data into C objects, so that you don't necessarily |
35
|
|
|
|
|
|
|
have to parse them yourself. See L for |
36
|
|
|
|
|
|
|
more on this. |
37
|
|
|
|
|
|
|
|
38
|
|
|
|
|
|
|
Designed to be used with L, these two |
39
|
|
|
|
|
|
|
modules aim to make it easy for anyone writing search engines based on |
40
|
|
|
|
|
|
|
Plucene. |
41
|
|
|
|
|
|
|
|
42
|
|
|
|
|
|
|
=head1 METHODS |
43
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
=head2 new |
45
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
my $indexer = Plucene::SearchEngine::Index->new( |
47
|
|
|
|
|
|
|
dir => "/var/plucene/foo", |
48
|
|
|
|
|
|
|
analyzer => "Plucene::Analysis::SimpleAnalyzer", |
49
|
|
|
|
|
|
|
); |
50
|
|
|
|
|
|
|
|
51
|
|
|
|
|
|
|
This creates a new indexer; you must specify the directory to contain |
52
|
|
|
|
|
|
|
the index, and you may specify an analyzer to tokenize the data. |
53
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
=cut |
55
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
sub new { |
57
|
1
|
|
|
1
|
1
|
409
|
my ($class, %args) = @_; |
58
|
1
|
50
|
|
|
|
8
|
croak("No directory given!") unless $args{dir}; |
59
|
1
|
|
|
|
|
6
|
my $self = bless { |
60
|
|
|
|
|
|
|
analyzer => "Plucene::Analysis::SimpleAnalyzer", |
61
|
|
|
|
|
|
|
%args |
62
|
|
|
|
|
|
|
}, $class; |
63
|
1
|
50
|
|
|
|
13
|
$self->{analyzer}->require |
64
|
|
|
|
|
|
|
or die "Couldn't require analyzer: $self->{analyzer}"; |
65
|
1
|
|
|
|
|
12141
|
return $self; |
66
|
|
|
|
|
|
|
} |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
=head2 index |
69
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
This adds a C to the index. |
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
=cut |
73
|
|
|
|
|
|
|
|
74
|
|
|
|
|
|
|
sub index { |
75
|
1
|
|
|
1
|
1
|
343
|
my ($self, $doc) = @_; |
76
|
1
|
|
|
|
|
56
|
$self->_writer->add_document($doc); |
77
|
|
|
|
|
|
|
} |
78
|
|
|
|
|
|
|
|
79
|
|
|
|
|
|
|
sub _writer { |
80
|
1
|
|
|
1
|
|
2
|
my $self = shift; |
81
|
1
|
50
|
|
|
|
18
|
return Plucene::Index::Writer->new( |
82
|
|
|
|
|
|
|
$self->{dir}, |
83
|
|
|
|
|
|
|
$self->{analyzer}->new, |
84
|
|
|
|
|
|
|
-e catfile($self->{dir}, "segments") ? 0 : 1 |
85
|
|
|
|
|
|
|
); |
86
|
|
|
|
|
|
|
} |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
=head1 Document Frontends and Backends |
89
|
|
|
|
|
|
|
|
90
|
|
|
|
|
|
|
So far so good, but how do you create these C? You |
91
|
|
|
|
|
|
|
can, of course, do so manually, but the easiest way is to use the |
92
|
|
|
|
|
|
|
supplied C or |
93
|
|
|
|
|
|
|
C modules. |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
These two modules are frontends which gather metadata about a file or |
96
|
|
|
|
|
|
|
URL and then hand the data off to one of the backend modules - there are |
97
|
|
|
|
|
|
|
backends supplied for PDF, HTML and plain text files. These in turn |
98
|
|
|
|
|
|
|
return a list of documents found in the file or URL. In most cases, |
99
|
|
|
|
|
|
|
there'll only be one document, but, for instance, a Unix mbox should |
100
|
|
|
|
|
|
|
return an object for each email in the box. These objects can be turned |
101
|
|
|
|
|
|
|
into C objects by calling the C method on |
102
|
|
|
|
|
|
|
them. This isn't done by default because you may wish to mess with the |
103
|
|
|
|
|
|
|
hash yourself, or serialize it, or whatever. |
104
|
|
|
|
|
|
|
|
105
|
|
|
|
|
|
|
=head2 Creating your own backend |
106
|
|
|
|
|
|
|
|
107
|
|
|
|
|
|
|
If you want to handle a different type of file, it's relatively easy to |
108
|
|
|
|
|
|
|
do. All you need to do is create a module called |
109
|
|
|
|
|
|
|
C; this should inherit from |
110
|
|
|
|
|
|
|
C and supply a |
111
|
|
|
|
|
|
|
C method. It should also call the |
112
|
|
|
|
|
|
|
C method to state which MIME types and file extensions |
113
|
|
|
|
|
|
|
it can handle. |
114
|
|
|
|
|
|
|
|
115
|
|
|
|
|
|
|
For instance, suppose we want to create a backend which grabs metadata |
116
|
|
|
|
|
|
|
from an image and indexes that. (Not unlike |
117
|
|
|
|
|
|
|
L...) We'd start off like this: |
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
package Plucene::SearchEngine::Index::Image; |
120
|
|
|
|
|
|
|
use strict; |
121
|
|
|
|
|
|
|
use warnings; |
122
|
|
|
|
|
|
|
use base 'Plucene::SearchEngine::Index::Base'; |
123
|
|
|
|
|
|
|
use Image::Info; |
124
|
|
|
|
|
|
|
|
125
|
|
|
|
|
|
|
Now we register the mime types and file extensions we can handle: |
126
|
|
|
|
|
|
|
|
127
|
|
|
|
|
|
|
__PACKAGE__->register_handler(qw( |
128
|
|
|
|
|
|
|
image/bmp .bmp |
129
|
|
|
|
|
|
|
image/gif .gif |
130
|
|
|
|
|
|
|
image/jpeg .jpeg .jpg .jpe |
131
|
|
|
|
|
|
|
... |
132
|
|
|
|
|
|
|
)); |
133
|
|
|
|
|
|
|
|
134
|
|
|
|
|
|
|
And our C method will call C for |
135
|
|
|
|
|
|
|
each bit of metadata it can find: |
136
|
|
|
|
|
|
|
|
137
|
|
|
|
|
|
|
sub gather_data_from_file { |
138
|
|
|
|
|
|
|
my ($self, $filename) = @_; |
139
|
|
|
|
|
|
|
my $info = image_info($filename); |
140
|
|
|
|
|
|
|
return if $info->{error}; |
141
|
|
|
|
|
|
|
$self->add_data("size", "UnStored", scalar html_dim($info)); |
142
|
|
|
|
|
|
|
$self->add_data("text", "UnStored", $info->{Comment}); |
143
|
|
|
|
|
|
|
$self->add_data("subtype", "UnStored", $info->{file_ext}); |
144
|
|
|
|
|
|
|
$self->add_data("created", "Date", Time::Piece->new( |
145
|
|
|
|
|
|
|
str2time($info->{LastModificationTime}))); |
146
|
|
|
|
|
|
|
} |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
See L for an explanation of C. |
149
|
|
|
|
|
|
|
|
150
|
|
|
|
|
|
|
Beceause C uses a plugin architecture, |
151
|
|
|
|
|
|
|
once this module is installed, it will automatically be called upon to |
152
|
|
|
|
|
|
|
handle those image types it can deal with, without any additional action |
153
|
|
|
|
|
|
|
by the user. |
154
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
=head2 Creating your own frontend |
156
|
|
|
|
|
|
|
|
157
|
|
|
|
|
|
|
For certain types of data, such as emails, news articles, or instant |
158
|
|
|
|
|
|
|
messages, you may not want to use the file or URL frontends. |
159
|
|
|
|
|
|
|
Alternatively, if you have a simple piece of data which isn't |
160
|
|
|
|
|
|
|
file-based, you may just want to do everything yourself. Even then, |
161
|
|
|
|
|
|
|
C can help you to create |
162
|
|
|
|
|
|
|
C - just inherit from it, and use C to add |
163
|
|
|
|
|
|
|
fields to the document in your C method. See |
164
|
|
|
|
|
|
|
L for more details. |
165
|
|
|
|
|
|
|
|
166
|
|
|
|
|
|
|
=head1 SEE ALSO |
167
|
|
|
|
|
|
|
|
168
|
|
|
|
|
|
|
L, |
169
|
|
|
|
|
|
|
L, |
170
|
|
|
|
|
|
|
L, L, |
171
|
|
|
|
|
|
|
L. |
172
|
|
|
|
|
|
|
|
173
|
|
|
|
|
|
|
=head1 AUTHOR |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
Simon Cozens C. |
176
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
This library is free software; you can redistribute it and/or modify it |
178
|
|
|
|
|
|
|
under the same terms as Perl itself. |
179
|
|
|
|
|
|
|
|
180
|
|
|
|
|
|
|
=cut |
181
|
|
|
|
|
|
|
|
182
|
|
|
|
|
|
|
|
183
|
|
|
|
|
|
|
1; |